Answer: ChatGPT uses automated tools, such as an internal version of the /moderations API, to detect content (prompts, completions, uploads) that may be harmful or violate Usage Policies. It also accepts human reports and uses a combination of automated systems and a trained team of experts to review them. Status: Confirmed Entry: tool/chatgpt/permissions-and-security/detect-problematic-content Revision: 1 Digest: sha256:a218f1e503ca099b (changes whenever the content changes) Timeline: Created: 2026-11-13T03:15:57Z Last changed: 2026-11-13T03:15:57Z Last verified: 2026-11-20T11:57:03Z Next check: 2026-11-21T11:57:03Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Source | As of detection method | automated tools, such as an internal version of our /moderations API | [1] | 2026-11-20 content types detected | prompts, completions, uploads | [1] | 2026-11-20 action on problematic content | warn you that your content may violate our usage policies or block the model from responding to your prompt | [1] | 2026-11-20 human reports review | a combination of automated systems and a trained team of experts | [1] | 2026-11-20 Sources: [1] OpenAI. "How we identify problematic content on our services for individuals". https://help.openai.com/en/articles/8940831-how-we-identify-problematic-content-on-our-services-for-individuals Role: primary. Accessed: 2026-11-20T11:57:03Z Retrieved via: DeepSeek Fetch API (Markdown conversion; the original server response was not captured) Quote: "We use automated tools, such as an internal version of our /moderations API, to detect content (prompts, completions, uploads) that may be harmful or violate our Usage Policies." (Supports: detection method) Quote: "to detect content (prompts, completions, uploads) that may be harmful or violate our Usage Policies" (Supports: content types detected) Quote: "we’ll typically either warn you that your content may violate our usage policies or block the model from responding to your prompt" (Supports: action on problematic content) Quote: "We use a combination of automated systems and a trained team of experts to review these reports." (Supports: human reports review) Snapshot SHA-256: bbd008b48348fdafb44a3f4a9ecd5fbebc4e0c6ace7a480a0fbb304900b0b0a2 (captured 2026-11-13T03:15:58Z) Verification: Method: Automated extraction from [1] with deepseek-flash (JSON output, thinking disabled). Every value was checked by code to appear verbatim in a quoted passage of [1]. [1] was retrieved through the DeepSeek Fetch API because a direct request did not return usable content. On this check the text of [1] was unchanged since 2026-11-13T03:15:57Z, so the values verified then were kept. Cross-check: not-available at 2026-11-20T11:57:03Z. No independent second source is configured for this entry Last successful check: 2026-11-20T11:57:03Z Last attempt: 2026-11-20T11:57:03Z Notes and limitations: Extracted automatically from a single official source (section "ChatGPT and ChatGPT Images"). Related: https://latestn.com/tool/chatgpt/permissions-and-security/disable-sharing-existing-access What happens to existing shared access in ChatGPT Space when an admin disables sharing? https://latestn.com/tool/chatgpt/permissions-and-security/dot-action-review-approval How does action review decide whether my dot needs my approval before acting? https://latestn.com/tool/chatgpt/permissions-and-security/dot-approval-link What happens when my dot needs approval for a request? https://latestn.com/tool/chatgpt/permissions-and-security/control-website-access-2 How do I control which websites the ChatGPT browser extension can access? https://latestn.com/tool/chatgpt/permissions-and-security/create-change-account-pin How do I create or change my account PIN for locked chats in ChatGPT? https://latestn.com/tool/chatgpt/permissions-and-security/definition-of-political-campaigning How does OpenAI define political campaigning in ChatGPT's usage policies? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-13T03:15:57Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/chatgpt/permissions-and-security/detect-problematic-content/latest Challenge this entry: Open challenges: None How: If a value here is wrong or out of date, send what is wrong and a link to evidence. Challenges are queued and checked against the evidence; they do not change the entry by themselves. If the entry changes as a result, the change log says so. Form: https://latestn.com/dispute/tool/chatgpt/permissions-and-security/detect-problematic-content POST: https://latestn.com/dispute/tool/chatgpt/permissions-and-security/detect-problematic-content with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/chatgpt/permissions-and-security/detect-problematic-content?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). How does ChatGPT detect problematic content in prompts and completions? Revision 1, last verified 2026-11-20T11:57:03Z. https://latestn.com/tool/chatgpt/permissions-and-security/detect-problematic-content (accessed 2026-11-21T10:01:41Z). @misc{LatestN_tool_chatgpt_permissions_and_security_detect_problematic_content_r1, title = {How does ChatGPT detect problematic content in prompts and completions?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/chatgpt/permissions-and-security/detect-problematic-content}}, note = {Revision 1; digest sha256:a218f1e503ca099b; last verified 2026-11-20T11:57:03Z; accessed 2026-11-21T10:01:41Z} } Reuse: License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/) Attribution: Copying and adapting are allowed, including commercially, if you credit LatestN and link to the entry you used. Each entry ends with a ready-made citation. URL: https://latestn.com/tool/chatgpt/permissions-and-security/detect-problematic-content